Analysis of Influencing Factors on Shale Gas Well Productivity based on Random Forest

Yu Zhou; Zaixun Gu; Changyu He; Bing Yang; Jian Xiong

doi:10.54097/adq74s58

Authors

Yu Zhou
Zaixun Gu
Changyu He
Bing Yang
Jian Xiong

DOI:

https://doi.org/10.54097/adq74s58

Keywords:

Random Forest, Shale Gas, Factor Analysis, Capacity Forecasting

Abstract

The objective of this study is to examine the primary controlling factors influencing shale gas well production. To this end, the research employs a random forest model to analyze the production capacity influencing factors based on the production data of a shale gas field in the Sichuan Basin. First, the process of analyzing the production capacity influencing factors based on the random forest model is designed for the high-dimensional, large samples, and multiple covariance characteristics of shale gas production data. The significance of each feature was evaluated through the impurity method and the out-of-bag data method, which identified the pivotal factors influencing the production capacity of gas wells across different time periods. To eliminate the potential influence of production enhancement measures on data analysis, this study further screened the core geological and engineering parameters and evaluated their significance. This research not only offers theoretical insights into shale gas exploration and development but also provides a reference point for predicting the production capacity of shale gas fields with com-parable geological characteristics.

Downloads

Download data is not yet available.

References

[1] Chen Gengsheng, Shi Xuewen, Liu Yong, et al. New understanding of the controlling factors of deep shale gas enrichment in the Wufeng Formation-Longmaxi Formation in the southern Sichuan Basin [J]. Natural Gas Industry, 2024, 44(01): 58-71.

[2] Li Yuegang, Zhou Anfu, Xie Wei, et al. Lithofacies division and main controlling factors of reservoir development of shale in the Wufeng Formation-Longyi_1 submember in Luzhou area, southern Sichuan Basin [J]. Natural Gas Industry, 2022, 42(08): 112-123.

[3] Shu Honglin, He Fangyu, Li Jilin, et al. Geological characteristics and favorable exploration areas of deep shale in the Wufeng Formation-Longmaxi Formation in Da'an Block, Sichuan Basin [J]. Natural Gas Industry, 2023, 43(06): 30-43.

[4] Huang Minghua, Zheng Wen, Zheng Hui. Key technologies and applications in shale gas extraction process[J]. Contemporary Chemical Research, 2024, (08): 83-85. DOI: 10.20087/j.cnki.1672-8114. 2024. 08. 026.

[5] Zhang Haijie, Mou Yongcheng, Han Guisheng, et al. Characteristics and distribution law of shale gas reservoirs in Upper Ordovician Wufeng Formation-Lower Silurian Longmaxi Formation in Dazu area, Sichuan Basin[J]. Natural Gas Exploration and Development, 2024, 47(04): 28-37.

[6] Hu Xi, Li Yading, Dan Xia, et al. Geological characteristics and main controlling factors of shale gas enrichment and high production in the Wufeng Formation-Longyi_1 sub-member in Luzhou area, Sichuan Basin [J]. Natural Gas Technology and Economy, 2024, 18(04): 15-23.

[7] Qing Peng. Analysis of dynamic characteristics and main controlling factors of deep shale gas [J]. Petrochemical Technology, 2024, 31(07): 238-240.

[8] Wang Wenjie, Ma Jinsha, Gao Qian, et al. Variable screening method based on random forest variable importance score and its application in tumor typing diagnosis [J]. Chinese Journal of Disease Control and Prevention, 2023, 27(3): 274-280.

[9] Cao Taoyun. Study on variable importance based on random forest [J]. Statistics and Decision, 2022(4): 4.

[10] Janitza S, Strobl C, Boulesteix A L. An AUC-based permutation variable importance measure for random forests[J]. BMC bioinformatics, 2013, 14: 1-11.

[11] Rigatti S J. Random forest[J]. Journal of Insurance Medicine, 2017, 47(1): 31-39.

[12] Nicodemus K K. On the stability and ranking of predictors from random forest variable importance measures[J]. Briefings in bioinformatics, 2011, 12(4): 369-373.

[13] Jiao Yanan, Ma Jie. An improved MEP decision tree pruning algorithm[J]. Journal of Hebei University of Technology, 2019, 48(6): 24-29.

[14] Ma Yankun, Guo Chaoyou, Xu Zhe. Research on fault diagnosis method of marine diesel engine based on random forest[J]. Ship Electronic Engineering, 2020, 40(11): 128-131.

[15] Xu Xiaoyang, Ji Zhicheng. Research on modeling of selective ensemble extreme learning machine classifier[J]. Computer Applications and Software, 2016, 33(9): 279-283.

[16] Wang Lijia, Zhu Zhengwei, Zhu Yanping, et al. A multi-strategy multi-objective reinforcement learning algorithm based on voting method[J]. Computer Simulation, 2023, 40(04): 341-345+406.

[17] Zhang Beibei, Hu Min. Customer satisfaction prediction based on grid search improved random forest[J]. Journal of Beijing Information Science and Technology University (Natural Science Edition), 2021, 36(4): 50-53.

[18] Li Zhengliang, Peng Sisi, Wang Tao. Surrogate model sequence sampling method based on k-fold cross validation[J]. Chinese Journal of Computational Mechanics, 2022, 39(2): 244-9.